Biogeographic Effects of the Closing Central American Seaway on Benthic Foraminifera of Venezuela

Studies of the effects of the Central American Seaway’s closure on microfossil taxa and stable isotopes of foraminifera previously addressed the timing of emplacement of the Caribbean–tropical eastern Pacific sill, changes in marine nutrients, evolutionary and paleobiogeographic events associated with closure, and rates of uplift in Central America. However, there have been no quantitative studies of the transisthmian divergence of whole biotas. Here we assess biogeographic effects of the closing Central American Seaway by comparing middle Miocene to Pliocene, neritic benthic foraminiferal assemblages from Falcón State, Venezuela, to those of coeval deposits with similar paleoenvironments from Panama, Costa Rica, and Ecuador. Benthic foraminiferal assemblages from the Urumaco, Codore, Caujarao, and La Vela formations of Venezuela’s central coast were compared to other Caribbean and tropical eastern Pacific assemblages with Simpson’s, Dice, and Jaccard similarity coefficients. Results of the Simpson’s Coefficient indicate that the Caribbean inner neritic faunas of Venezuela, Costa Rica, and Panama became significantly more similar from the late Miocene to Pliocene, as mixing with Pacific waters was reduced and modern oceanic conditions of the Caribbean Sea were established. Twenty million years ago, according to all paleoceanographic evidence, the Central American Seaway connected Caribbean and tropical eastern Pacific waters. Over the next 15 My, the movement of tectonic plates and uplift caused shoaling of the seaway to depths that restricted deepwater flow between the tropical Atlantic and Pacific until complete emergence of Central America approximately 4 Ma. Paleontological research on evolution and biogeographic changes resulting from seaway closure has focused on shallow-water invertebrate and protist fossils in land-based deposits, and deepsea or open-ocean, surface-water protists preserved in deepsea cores. Here we briefly review the physical cutoff of Caribbean from tropical Pacific waters and the microfossil response, then quanitatively compare whole tropical American benthic foraminiferal assemblages through the late Miocene and Pliocene. Uplift of the Isthmus of Panama and Paleoceanographic Changes The southern Central American arc on the southwestern edge of the oceanic Caribbean plate has been moving generally eastward since at least 65 Ma, to finally collide with continental South America. Coates et al. (2004) used the regional, land-based microfossil record of paleobathymetry and regional unconformities of southern Central America and northwestern South America to put the time of their collision at approximately 15–13 Ma, within the range of all estimates other than Farris et al. (2011), who suggested 20 Ma. After intial collision, marine basins from BULLETIN OF MARINE SCIENCE. VOL 89, NO 3. 2013 922 northwestern Colombia to southeastern Costa Rica generally shallowed to neritic depths. A regional unconformity suggests collision was completed by approximately 7 Ma (Duque-Caro 1990, Collins et al. 1995, 1996b, Coates et al. 2003, 2004), although there were localized uplift events such as that caused by the subduction of the Cocos Ridge beneath Central America (Collins et al. 1995). The progressive closure of the Central American Seaway and resulting oceanographic changes have been studied primarily with proxies of physical oceanography and general circulation models. As of 8 Ma, a reduction of eastward-flowing Pacific bottom water into the tropical Atlantic was indicated by low Nd and high Pb radioisotopes at North Atlantic sites (Frank et al. 1999, Reynolds et al. 1999). Throughout the late Miocene, the Caribbean was poorly ventilated, as seen from amounts of foraminifer shells and δ13C (Bickert et al. 2004). About 8 Ma mesotrophic conditions with little ventilation of bottom waters and low current velocity were indicated by benthic foraminiferal proxies (Jain and Collins 2007). Between 8 and 7 Ma, nutrient-rich bottom waters flowing into the Atlantic were reduced, based on diverging Caribbean–eastern Pacific δ13C values (Billups 2002, Spezzaferri et al. 2002). By about 6 Ma, benthic foraminiferal faunas show that bottom-water masses on either side of the isthmian deep-water sill had diverged, because Antarctic Deep Water was replaced in the Caribbean by North Atlantic Deep Water and in the Pacific by Pacific Deep Water (McDougall 1996). Seaway closure decreased Atlantic primary productivity and reduced the inflow of high-nutrient, subsurface Pacific water, according to a climate-ocean model of differing sill depths and vertical diffusivity (Schneider and Schmittner 2006). For the eastern equatorial Pacific, the same model predicted increasing productivity, agreeing with foraminiferal δ13C gradients that suggested an increase in surface-water nutrient concentrations during the last 5 My (Cannariato and Ravelo 1997). Foraminiferal proxies show decreased paleoproductivity after about 4 Ma (Bornmalm et al. 1999, Jain and Collins 2007, Jain et al. 2007). About 3 Ma, Caribbean surface waters became oligotrophic and the thermocline/nutricline deepened, based on calcareous nannofossil assemblages (Kameo 2002). Most paleoceanographic proxies agree that the Central American Seaway closed completely about 4 Ma (although Pleistocene sea-level rises may have breached the isthmus; Keller et al. 1989). Whereas recent lithologic studies of the Central American volcanic arc proposed seaway closure by 15 Ma (Farris et al. 2011, Montes et al. 2012), their direct evidence is for the geological collision of elements of the Isthmus of Panama rather than the vertical changes in land vs sea that directly caused seaway closure (Coates and Stallard 2013). Studies of foraminiferal δ18O in deepsea cores indicated that the modern Caribbean–eastern Pacific salinity contrast of 2 began to develop around 4.5 Ma (Keigwin 1982a, Haug et al. 2001) as a result of higher evaporation in the Caribbean and reduced mixing with the Pacific. Although Molnar (2008) suggested that an early Pliocene change from a permanent El Niñolike state to the modern El Niño-Southern Oscillation state could also explain transisthmian divergence in δ18O, the proposed changeover has not been explained. In any case, after accounting for changing salinities, δ18O also reflects North Atlantic surface warming as shown by Mg/Ca ratios that indicated an increase of 2 °C in the central Caribbean at about 4.5 Ma (Groeneveld et al. 2008). smith et al.: biogeography of venezuelan foraminifera 923 Efffects of Changes in Seaway Depth on Microfossils During the Paleogene, the gap between Central America and South America was bordered by a shallow carbonate platform extending north to central Panama (Cole 1952, 1957). In the early Eocene, bathyal to abyssal facies existed in present central to eastern Panama, as indicated by mollusks (Woodring 1970), foraminifera, and radiolarians (Bandy 1970, Bandy and Casey 1973). Other deeper, bathyal facies of southern Central America (Collins 2007) possibly resulted from local basin subsidence or down-dropped grabens. By late-early to early-middle Miocene, the southern Central American volcanic arc became more emergent, with foraminiferal faunas that shallowed from lower-middle bathyal to outer neritic depths (Collins et al. 1995, Coates et al. 2004), but a lower bathyal-abyssal seaway in eastern Panama–southwest Colombia still existed as indicated by benthic foraminifera (Duque-Caro 1990). Late-middle to early-late Miocene foraminifera of the Gatun Formation (Cushman 1918, Collins et al. 1996b), central Panama, were inner-middle neritic, indicating a closed Panama Canal strait. However, bathyal foraminifera of the overlying Chagres Formation suggest re-opening of the strait about 6 Ma (Collins et al. 1996b). Microfossil paleobiogeography agrees with the early-middle Miocene history of Central American Seaway depths. In the early Miocene, radiolarian taxa that had been distributed across the tropical Pacific and Atlantic gradually declined to virtual disappearance while their Pacific counterparts remained (Maurrasse 1979). Earlymiddle Miocene, bathyal benthic foraminiferal faunas of Venezuela and Ecuador were very similar with a broad tropical distribution (Cushman 1929, Hasson and Fischer 1986, Whittaker 1988), although Duque-Caro (1990) noted that Caribbean and Pacific faunas of Colombia were different. By middle Pliocene, Ecuador foraminifera had “much closer affinities” to those of California than the Caribbean, resulting from a Caribbean-Pacific barrier by this time (Hasson and Fischer 1986). The late Miocene was a time of great evolutionary and paleogeographic change in Caribbean-tropical eastern Pacific microfossils. Origination in common, shallow-water benthic foraminifera of southern Central America increased in the late Miocene (Collins et al. 1996a). Late Miocene outer neritic foraminiferal faunas of coastal Ecuador had less similarity with the Caribbean than did shallower, middle neritic faunas of the same formation, possibly reflecting development of tropical Pacific-Caribbean endemism with the rising sill (Collins 2006). In the deep eastern equatorial Pacific, amounts of middle Miocene, transported neritic species with Caribean affinities decreased in the late Miocene (McDougall 1985). The diversity of deepsea Caribbean foraminifera peaked around 8 Ma during high paleoproductivity and declined thereafter (Jain et al. 2007), although the diversity of shallow-water taxa of southern Central America has increased until today (Buzas et al. 2002). Complete closure of the Central American Seaway about 4 Ma apparently had less effect on benthic microfossil evolution than did late Miocene events (Collins et al. 1996a). The occurrence in both the tropical Atlantic and eastern Pacific of the planktic foraminifera Globorotalia pertenuis Beard, 1969 and Globorotalia miocenica Palmer, 1945 (Keigwin 1982b) that originated about 3.5 Ma confirms surfacewater exchange across the isthmus. Also occurring in planktic foraminifera at about 3.5 Ma are discordant patterns between the Atlantic and Indo-Pacific oceans in the coiling direction of Pulleniatina (Saito 1976), and an abrupt increase in the average BULLETIN OF MARINE SCIENCE. VOL 89, NO 3. 2013 924 number of supplementary apertures in the Sphaeroidinella lineage (Kucera 1998); both events are ascribed to complete seaway closure.